Mar 2019 – May 2020 · 1 yr 2 mos · Bengaluru Area, India

• Work as the lead SRE in Uber Marketplace Platform(MP) SRE team. Joined as the first SRE and successfully bootstrapped a team of 6+ members. Involved in the hiring, training, mentoring of the team members from the inception.
• Implemented SLA/SLO for Marketplace Platform Services.
• Platform level and Application level Capacity testing and planning
• Improving observability of Marketplace Platform.

Sep 2016 – Mar 2019 · 2 yrs 6 mos · Bengaluru Area, India

• Managing 100PB+ Apache Hadoop Clusters and Presto clusters at Linkedin.
• Responsible for Hadoop Performance, reliability, capacity planning and Monitoring
• Support other Hadoop Echo systems such as Apache Hive, Pig, Spark,etc
• Automation of various tasks including cluster management tools, capacity planning tool etc.
• Working with multiple configuration Management systems such as bcf2, salt etc.
• Active role in planning and implementation of Preso at Linkedin.

Jul 2014 – Sep 2016 · 2 yrs 2 mos · Bangalore

• Primary ownership of Grid Platform and Data Streaming/Messaging platform
• o Have deep level operational understanding of HDFS and YARN
• o Hands on experience with other Grid components Hbase,Zookeeper,Oozie and Falcon
• o Work closely with platform dev team, actively takes part in discussions on improving cluster stability, adding new features, configuration tunings, operability improvements, ongoing issues etc.
• o Implemented Namenode and ResourceManager High availability
• o Implemented cgroup in yarn to control and govern CPU resource utilization across multiple tasks
• o Capacity Planning and cluster augmentation
• o Troubleshooting issues related to hadoop ecosystems.
• o Troubleshooting issues related to Linux OS, Hardware etc
• o Graphing of Hadoop jmx metrics using Grafana and Graphite.
• o Own inmobi messaging/streaming platform, that constitute of scribe messaging
• service and conduit
• Hadoop Upgrades
• o Planned and executed Cloudera(CDH4) to HortonWorks(HDP) 2.2.4 ( Hadoop 2.6.0) upgrade. Worked closely with dev, QA, other stakeholders and were able to complete this task without any surprises.
• o Takes care of minor version upgrades (patches, bug fixes) and configuration changes across grid clusters
• Hadoop Monitoring
• o Graphing of hdfs and yarn jms metrics using graphite and grafana.
• o Graphing of system metrics using collectd, graphite and grafana
• o Use Nagios for sending oncall alerts
• HDFS/yarn Tuning
• Very good understanding of HDFS and YARN configurations

Dec 2007 – Jul 2014 · 6 yrs 7 mos · Bengaluru Area, India

• Managing Yahoo advertising applications using the Yahoo Hadoop eco system
• Work with cross-functional team to onboard new pipelines
• Capacity Planning and conducting Load test on hardware.
• Primary ownership of Yahoo Image search crawlers and thumbnails
• Prepared and executed migration plan for upgrades like FreeBSD to RHEL Migration, PHP 4 to PHP5 Migration, Migrating Legacy component to Yahoo Standard component.
• Configuration and management of RHEL hosts which serve Yahoo Image Search web traffic.
• Performance tuning for Image Search web servers and backend clusters running on RHEL.
• Creating automation tools for tasks which requires manual attention on daily basis.
• Implemented Business Continuity Plan for multiple components.

Mar 2004 – Nov 2006 · 2 yrs 8 mos